Benchmarking food consumption estimates from imports data and net food disposable data with dietary survey data

Abstract

Food consumption data obtained through dietary surveys is critical information for crafting effective health policies. However, conducting large-scale surveys on a regular basis can be expensive. In this work, we compare nutrient estimates derived from food supply proxy datasets created using: 1) imports data rescaled with import reliance metrics and 2) imports data combined with net food disposable data (NFD). The bulk nutrient distributions from the food supply proxy datasets are normalized with population estimates that align the nutrient supply per capita of the survey data and the food supply proxy data for a chosen macronutrient to compare the resulting nutrient estimates for different nutrients used as a baseline. The distribution of RDIs for generated plates using the combined imports and net food disposable data with energy per capita aligned to survey data was compared to see how individual intake from the survey compares to simulation. We found that using imports with NFD data as a food supply proxy dataset results in closer RDI values to the survey data compared to a food supply proxy dataset derived from rescaling imports only.